That system involves using a couple of different programs in sequence. I’ve distilled it down to one program called Flutini which is less flexible but easier to use and can show your tuning live as you play. It is free and open source and available here:
That page has a better description, screenshots and a tutorial if you are interested. (The program is only available for Windows. If someone is interested in doing a port to the Mac and/or Linux, let me know.)
Fascinating, and I can’t wait to try it! But notes are not binary switches, with only on and off values. There are different types of transitional tones, slides, taps, crans, etc… Wouldn’t these skew the values, and be recognized as off-pitch notes?
Flutini is based on Philip McLeod’s dissertation work at University of Otago, NZ, and much of what he did was to address the problem of note identification in messy real music. On top of that, Flutini does a number of things to eliminate transitional notes, etc. It filters out muddy notes, notes that only appear a few times, notes that have low volume and so on. And Flutini reports the average for a note based on many samples, so unless you’re doing lots of long slides, the average should not be greatly affected. (Taps, crans, rolls, etc., aren’t a big problem – the computer can listen and process much faster than your fingers can move!
I can see th latter two but…Idealists?? We tend more towards OSX and Linux I should think.
Certainly wishful thinking that Windows will ever get Vista right.
Somebody please get Flutini together for those of us who use Macs.
What I REALLY want is an iphone app that will allow you to record audio (lessons, sessions, workshops etc.) on the iphone, along with a slow downer/note and tuning analyzer program. The SDK is out there, and the apple app store will sell it for you, anyone?
I love the way you can adjust your real-time intonation as you play to reduce the horizontal bars to zero. Great for ear training to the quirks of a particular instrument.
Will there be a way to:
o Set the reference pitch to other than A440?
o Set a different temperament and/or custom intonation?
So now, flute players, makers and repairers, there are no excuses not to get into RTTA at the level that meets your needs. Mac owners can still use the Polygraph while we wait for someone to port Flutini for them.
If you haven’t tried RTTA yet, Flutini offers a very easy start and may well meet all your needs. If not, it’s then very easy to add the extra bits needed for the Polygraph. Give it a go!
And if you’ve fallen behind on the whole RTTA story, start off at:
Actually, I’ve already had another request to support A415 for baroque flutes. I don’t think changing the reference pitch is terribly difficult, so I’ll have a look at it. Changing the temperament is a little trickier (because, I have to admit, that I’m a lazy programmer).
If you are going to change the ref pitch, can you make it from 390Hz (old French Pitch) to about 460Hz (above British High Pitch). That should take care of all comers.
Is a possibility for the temperament to replace the vertical 0 line with a bent and twisted goalpost? That would show the response against the selected temperament and in relation to equal temperament at the same time.
I’ll likely make it 400-500 or something like that. I think this just shifts the whole scale a linear amount (I’ll check Graeme’s code to be sure, since I assume he knows what he’s doing ) so there’s no big deal to do any amount of offset.
Is a possibility for the temperament to replace the vertical 0 line with a bent and twisted goalpost? That would show the response against the selected temperament and in relation to equal temperament at the same time.
Hi Scott
It’s a log scale not linear (log2(pitch/440)*12.00) so pitch at 220 or at 880 will move it 12 semitones. You should be able to do this on the fly easily enough.
Let the user choose any pitch from 220 to 880 and they can use this to give the nominal fingered note for a transposing instrument like a Bb Irish flute F fife Eb whistle etc. This is how Polygraph v 2 is doing it although in v3 you’ll be able to enter the key of the instrument rather than needing to know the pitch in Hz
If you can make it down to 390, try to do so, or we’ll all be deafened by the whimperings from the serious early music set. Not the wuzzy 415 “baroque pitch” lot (heh heh), but the people who actually play accurate copies of flutes by Hotteterre, Chevalier etc that were pitched below 400Hz.
I’ve put a new version of Flutini out there which has options for modifying A440 and different temperaments. Right now it supports equal, Just-D and 3 temperaments that the user can input (and the program will save them for future use). I think I’ve got the A440 correct but I’m less sure about the temperaments, so if you can check it out and give me feedback it would be appreciated.
If there are other “standard” temperaments that I should include as options it is fairly trivial.
Hi Scott
The pitch settings are working OK but are back to front! ie you’re adding when you should be subtracting.
The temperament settings are not working properly at all. If I set it to justD first the figures for the accidentals shouldn’t be 0 but you’ve got Fnat at 0 G at -2 …Now if I play Fnat it gets assigned to both Fnat and G exactly the same amount of data out by the same amount. Seems it’s doing the temperament in semitones not cents…
I tried out Flutini today on an old clunker computer I keep in the workshop for tuning (running Tatsuaki Koroda’s Auto Tuner under Windows '98 ). Seems to run fine, which is really going to make retuning old flutes very easy!
We’ve probably all been assuming that fiddle players would prefer to play a just scale, but in an email today, (Prof) Neville Fletcher advises that this may not be a safe assumption:
“Measurements on violinists indicate that they tend to tune major thirds sharp – even sharp compared with the equal tempered value, which is itself sharp compared with a “just” 5:4 major third. This may perhaps be because they have tuned their strings to perfect 3:2 fifths, so that they tend to favour Pythagorean tunings.”
Now maybe “violinists” and “fiddle players” are not the same thing, at least in some issues. This is a matter we could investigate with RTTA, Tartini or Polygraph. If anyone has some spare time and a few recordings of solo fiddle players …
Just started playing with Flutini because I am trying to figure out some tuning changes on my flutes. I obviously need some flute-making advise!
On the F flute I just made it plays quite well (well, for a beginning making) in the first octave but goes REALLY flat on the top of the second octave. Now, I know that is to be expected with a cylindrical flute but this is far more than on the D flutes I am making and wondered if there is an obvious reason for it.
Here is the flutini result;
Any ideas on why I am seeing such a large difference? Do I need to adjust the C and D holes to play sharper and just play into them far more on the lower octave?
Could the embouchure have much to do with this? I have a 7/16 embouchure, which I prefer over the 3/8 but am wondering if it makes it harder to lip up the second octave…? When I play my D flute with 3/8 embouchure there is no issue on keeping the second octave the same as the first.
I have played around with the cork positioning and have it at about 18mm, which seems about the best spot, but am still having this issue.
We tend more towards OSX and Linux I should think.
) so there’s no big deal to do any amount of offset.